useless questions

[dfan]

Q 2: There are so many go skills that being good in some areas can compensate for being bad in others, and the game is so long that a single error rarely produces a loss, except at the highest levels. Still, I think that weaknesses are more of a drag than strengths pull you up. My guess is that it is hard to play more than 3-4 stones stronger than your weakest area, and easy to play 9 stones weaker than your strongest area.

I enthusiastically agree. The best way to get from 10k to 9k is to stop making 12k mistakes

[Tryphon]

Question 1:
Have you ever experienced a sudden jump in strength because you learned/understood one particular subject (e.g. efficient positional judgement, understanding when and how to sacrifice stones, how to avoid aji-keshi, when to ignore atari, when to tenuki, etc)?

I feel like I'm learning sabaki. It's a pleasure to see my opponent asking himself how to answer a move that didn't cost me anything, it was the opposite for so long...

Sounds more like you're learning probes to me

Maybe the two are coming together And bad aji too

[Toge]

IMO, 90% of your go strength is reading power. Many other skills are useful only in that they support your reading

The remaining 10% is positional judgment, concentration, stamina, etc.

- As 2kyu I read only very specific situations. Most of the skill is atomistic memorization. For example, making a safe extension from third line stone requires no thought; I already know that two spaces is the distance. Same goes to other useful shapes and tactics like capture races.

I think dan players exaggerate value of reading, because they have reached a point where memorization is no longer relevant. Memorization is very relevant to quickly get out of beginner class. I few simple principles did this to me.

[Magicwand]

I think dan players exaggerate value of reading, because they have reached a point where memorization is no longer relevant. Memorization is very relevant to quickly get out of beginner class. I few simple principles did this to me.

how many instance in your game memorization actually help?
few opening moves and josekies perhaps..
rest of them are reading.

there are saying: "memorize joseki and the forget them"
you will forget them when you are strong enough and not have to rely on your memorization.
after that your reading will free you from boring josekies and you can create your own joseki or sometimes prove that existing josekies are wrong.
in my opening game of go is 99% reading.

[Chew Terr]

how many instance in your game memorization actually help?

By memorization, I think Toge is referring to a lot of the tools that you use to read. For example, knowing that placement tesuji often exist in a shape like 'x', or recognizing dead eyeshape and such. If I could manage to memorize every tesuji in every book, at least I'd have a better chance of seeing them.

I think that a reasonable metaphor is furniture. The serious person may like the idea of handmade furniture, and reject the idea of getting anything at all from a store. He may spend days turning out beautiful chairs and tables, far above what could be found in the local Sears.

However, there are two other options, not one. The option akin to pure joseki memorization is going to the store and buying a table. And you're right, that table is going to look like crap. =)

The alternative that Toge seems to suggest is to use memorization to get to the point where reading is useful and viable. Sure, we're going to the store, but we're going to buy hammers and saws.

I know this metaphor's not perfect, but I stand by it. Memorization is far from a full approach, but there is a lot to be said for the memorization used for shape/tesuji recognition. If I see a group on the second line, I want my mind to inherently think 'don't crawl', just because I've seen the results often enough to recognize them as bad.

Caveat: Tesujis and such require reading, of course, to ensure that the tesuji will work. One should never be without the other. I'm just saying it's hard to find the tesuji if you don't recognize the shape in the first place.

[Toge]

how many instance in your game memorization actually help?
few opening moves and josekies perhaps..
rest of them are reading.

- Quite a lot:

* Opening principles
* Joseki
* Extensions
* Whole-board evaluation
* Types of connections
* Types of tesuji
* Capture race principles
* Dead shapes
* Probes
* Forcing moves

These on the top of my head. A 20k is not 20k because he can't read. He's 20k because he has no idea what to read. Everything that happens in a game for him is unexpected. Playing many games allows situations to repeat and in those occasions learning happens. You need to brune that tree of possibilities to a few relevant moves, which you can then read in-depth using relevant counter moves. That's what strong players are doing.

[Magicwand]

how many instance in your game memorization actually help?
few opening moves and josekies perhaps..
rest of them are reading.

- Quite a lot:

* Opening principles
* Joseki
* Extensions
* Whole-board evaluation
* Types of connections
* Types of tesuji
* Capture race principles
* Dead shapes
* Probes
* Forcing moves

These on the top of my head. A 20k is not 20k because he can't read. He's 20k because he has no idea what to read. Everything that happens in a game for him is unexpected. Playing many games allows situations to repeat and in those occasions learning happens. You need to brune that tree of possibilities to a few relevant moves, which you can then read in-depth using relevant counter moves. That's what strong players are doing.

* Opening principles = reading
* Joseki =memorization + reading.
* Extensions = reading
* Whole-board evaluation = reading
* Types of connections = reading
* Types of tesuji = reading
* Capture race principles = reading
* Dead shapes = there are only few dead shapes. getting to dead shape is reading.
* Probes = reading
* Forcing moves = reading

get my point?

[Toge]

* Opening principles = reading
* Joseki =memorization + reading.
* Extensions = reading
* Whole-board evaluation = reading
* Types of connections = reading
* Types of tesuji = reading
* Capture race principles = reading
* Dead shapes = there are only few dead shapes. getting to dead shape is reading.
* Probes = reading
* Forcing moves = reading

get my point?

- That's from your point of view. After taking all atomistic knowledge you have for granted.

Analogy:
To type a sentence, you have to understand what each word means. Typing a sentence here is production. Arranging words in correct order to have meaningful result. In order to type a sentence, you don't need to invent any words. They're all in dictionary. You've heard all the words you're using somewhere before. That's knowledge. Same goes to game of Go: you hear the proverb "hane at the head of two stones is a good move" and the first move you'll consider in situation like that is hane. This knowledge makes weak players stronger.

[Magicwand]

Same goes to game of Go: you hear the proverb "hane at the head of two stones is a good move" and the first move you'll consider in situation like that is hane. This knowledge makes weak players stronger.

i play hane at the head of two stone because i read what happens after not because i memorized it.

[entropi]

Going to one extreme:

A program which memorized all the pro games, all known joseki, all standard shapes and their weaknesses, all proverbs, all strategic concepts, etc... But it is not able to read one move further. Such a program cannot be strong because it will be destroyed with the first slighest
deviation from a joseki, for example.

To the other extreme:

A computer program just being able to apply the rules and search the move-tree in depth. Despite the enormous reading ability, such a computer program cannot be strong because it lacks two vital skills:

1- It does not know what to read: This can only be irrelevant if the reading power is infinite. But for being realistic, we have to assume it is not infinite and it cannot search the complete move tree.

2- Even if it reads the correct move and every possible follow-up, it still does not know how to assess the result of his reading.


My conclusion:
One without the other (memorization and reading) is useless.

[HermanHiddema]

Analogy:
To type a sentence, you have to understand what each word means. Typing a sentence here is production. Arranging words in correct order to have meaningful result. In order to type a sentence, you don't need to invent any words. They're all in dictionary. You've heard all the words you're using somewhere before. That's knowledge. Same goes to game of Go: you hear the proverb "hane at the head of two stones is a good move" and the first move you'll consider in situation like that is hane. This knowledge makes weak players stronger.

I think that in general, the language analogy is very good. Go is a lot like talking. It is no coincidence that go is also called handtalk sometimes

Learning to play go is like learning a new language.

Learning shapes, tesuji, probes is like learning words, expanding your vocabulary.
Going over pro games is like reading books, seeing how words are used by native speakers.
Learning joseki is like learning set phrases, like "Can you tell me the time?" or "How much for those apples?"
Doing life & death is like conversation practice: "At the hotel", "At the airport", "In the shop".
Standard openings are like standard conversation starters, like "How was work today?"

But playing a game is like having an actual conversation. You still have to tie all of what you've learned together. Which words are appropriate to express what thought? What response is appropriate to which question? Is this consistent with the rest of what I've been saying? Even though you're using the same words, every conversation is unique.

Having an actual conversation means you have to think about what you are saying. That is what reading is in go. All those memorized components will make your conversation better, easier and faster, but you do have to think about how to apply them.

[topazg]

i play hane at the head of two stone because i read what happens after not because i memorized it.

I think you're slightly missing the point perhaps. For example, in the below diagram:

$$Wc Move 1
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . 1 , B . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . 2 . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------

Click Here To Show Diagram Code

[go]$$Wc Move 1
$$ ---------------------------------------
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . 1 , B . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . 2 . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . , . . . . . , . . . . . , . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ | . . . . . . . . . . . . . . . . . . . |
$$ ---------------------------------------[/go]

How often have you even bothered to read variations starting with Black 2?

It is memorising (call it experience if you'd rather) that gives us a valuable initial selection of moves to bother reading further.

EDIT: Also, I can think of a number of Malkovich game moves where you've basically said "Don't know if this works, haven't read it, but it looks right" or something like that

[topazg]

I think that in general, the language analogy is very good. Go is a lot like talking. It is no coincidence that go is also called handtalk sometimes

Learning to play go is like learning a new language.

Learning shapes, tesuji, probes is like learning words, expanding your vocabulary.
Going over pro games is like reading books, seeing how words are used by native speakers.
Learning joseki is like learning set phrases, like "Can you tell me the time?" or "How much for those apples?"
Doing life & death is like conversation practice: "At the hotel", "At the airport", "In the shop".
Standard openings are like standard conversation starters, like "How was work today?"

But playing a game is like having an actual conversation. You still have to tie all of what you've learned together. Which words are appropriate to express what thought? What response is appropriate to which question? Is this consistent with the rest of what I've been saying? Even though you're using the same words, every conversation is unique.

Having an actual conversation means you have to think about what you are saying. That is what reading is in go. All those memorized components will make your conversation better, easier and faster, but you do have to think about how to apply them.

This is awesome!! You've taken a generally reasonable analogy and made it a fabulous one. This needs recording somewhere on SL or something..

[Harleqin]

Same goes to game of Go: you hear the proverb "hane at the head of two stones is a good move" and the first move you'll consider in situation like that is hane. This knowledge makes weak players stronger.

I play hane at the head of two stone because I read what happens after, not because I memorized it.

And why do you even begin to read the hane? (Besides, I thought you did not like to read.)

[karaklis]

A 20k is not 20k because he can't read. He's 20k because he has no idea what to read.

Exactly that, and it has to be emphasized once more.

You can also put it the other way around: You can only get strong at reading if you know lots of shapes and tesuji.

To abstract it: Suppose you can read 100 positions. When you're 20k, there are about 10 eligible positions to play on every move. Then you will be able to read out two moves, your move, your opponent's move, that's all - reading depth: 2. When you're a dan player, there are 1-2 eligible positions. As for the others you *know* that these are bad. Assuming 1.5 positions per move you will be able to read out a move depth of 11-12.

Once I read an article about what professional chess players makes so strong. To find out about it, their brain activities were examined during play. It turned out that their long-term memory was much more active than that of ordinary players. This finding led to the conclusion that professional players read out/assess the game by accessing patterns from their long-term memory. This makes perfect sense for go as well, since go is even more related to pattern recognition than chess.

So to get stronger you need to internalize lots of patterns in your long-term memory.